 Hi, this is Dr. Don. I have a problem out of Chapter 6 about confidence intervals and it reads like this. We're given a random sample of 22 people. The mean commute time was 33.5 minutes and the standard deviation was 7.1 minutes, where to assume the population is normally distributed and they want us to use a t-distribution to construct a 99% confidence interval for the population mean and then find the margin of error of the mean. Okay, let's read this. They tell us t-distribution, but what if they had not said use a t? How would we know to use a t instead of a z? Well, we've got a random sample of 22, which is less than 30. So that's a clue that since it's less than 30, you should use the t. Over here we're given the x-bar, the mean commute time in the sample, and s, the standard deviation in the sample. We're not given the population standard deviation sigma. So that again points toward the t-distribution instead of the z. So let's bring up stat crunch now and solve this quickly. As we often do, we'll start with stat and t-stat because we're using the t-distribution. This is one sample and we're given a summary. So we bring up our dialog box. The sample mean is 33.5. The sample standard deviation s is 7.1. The sample size is 22. We're going to click on confidence interval for the mean and we point 0.99, 99%. I'm going to this time click and store the data in the data table, or the output in the data table. So we click that out of the way, and we've got this data summarized there, our mean, sample mean of 33.5, the standard error, degrees of freedom, and there's our lower limit and upper limit. If you remember, the upper limit minus the lower limit is the width of the interval. The margin, the error is half of the width and it's also equal to the mean minus the lower limit, or the upper limit minus the mean. So I'm going to use the data compute expression path here just to do it all in stat crunch. And I'm going to take the sample mean minus the lower limit, click OK, and I'm going to label this as E, our margin of error, and click compute, and it's 4.28, which rounds to 4.3, which is the answer they wanted. And of course we've got our lower limit, 29.2, upper limit, 37.8. So that's the answer to their problem. I hope this helps.